Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: Comparing sureg with reg

From   DE SOUZA Eric <>
To   "" <>
Subject   st: RE: Comparing sureg with reg
Date   Tue, 31 Jan 2012 17:28:01 +0100

As you say: " The 45 independent variables are the same for all three equations"
Estimating a seemingly unrelated regressions system in which the explanatory variables are the same for all equations is equivalent toe estimating each equation separately by OLS inasfar as the parameter estimates and standard errors are concerned.

Eric de Souza
College of Europe
Brugge (Bruges), Belgium

-----Original Message-----
From: [] On Behalf Of Schöler, Lisa
Sent: 31 January 2012 12:09
Subject: st: Comparing sureg with reg

Dear Statalist,

I have three equations with three different dependent variables and 45 independent variables. 
I ran a SUR model with the command -sureg-. The 45 independent variables are the same for all three equations.
The Breusch-Pagan test of independence shows that my dependent variables are not independent (p<.001).
When I compare the coefficients that I receive from -sureg- and -reg-, they do not differ at all. Only the standard errors differ, but only slightly.

Can anybody explain me what I am doing wrong here, I have no explanation why the coefficients do not differ at all? I know that this is the case when you only have dummy variables, but I have 30 dummy variables and 15 variables that are not dummies.


Lisa Schöler
Professur für BWL, insb. Electronic Commerce Goethe-Universität Frankfurt Grüneburgplatz 1
60629 Frankfurt a.M.
Phone: ++49 69 798 34632
Fax:   ++49 69 798 35001

Get serious about better marketing decisions - apply now!

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index